Monday, October 5, 2009

Some general principals that apply to bicycle mechanics:

Generally speaking, bicycle mechanics adjust things. We also lubricate things that should slide (cables) and things that should spin (hubs, bottom brackets, etc) and things that pivot (brakes and derailluers).

As far as adjusting things goes, there is a type of part that we adjust so it is not too tight and not too loose. Those parts would include the bottom bracket, hubs, headset and in the case of sidepull brakes, their pivots.

In the classic parts, as opposed to some of the newer sealed bearings, there are threaded parts that we adjust tighter or looser to the point where there is the least amount of pressure that doesn't result in play at the bearing or brake pivot. The mistake I see most often is when the locking characteristics achievable with threaded adjusters is not used properly. That is, the bearing may have been adjusted pretty well, but the adjustment is not locked in.

I will use a hub adjustment to illustrate the principle. Because of the difficulty of trying to convey this process with words alone, I will use as an example, a hub that has a hex shaped lock nut. That is the nut that sits next to the frame. First, I put the wheel horizontally in my vise. I clamp the vise onto the locknut and do not put it in too deeply because I need access to the cone. That is the part that contacts the bearings. The locknut and cone normally have a washer between them.

The first adjustment I make when the wheel is sitting like a big steering wheel in the vise is to lock down the bottom cone. Since the bottom locknut is held in the vise, I back the cone down into it and doing so locks the two together into that position. That principal is called "jamb nutting". It is the same principle used to adjust the valves on VW and other engines.

Once the bottom pair of lock nut and cone are working as a team, I move my attention to the top set of components. Remember that the wheel is sitting horizontally so there is a top set and the already locked in place bottom set. Next I will adjust the bearing so it will work as long as possible. (in another paragraph I will explain why there should be a tiny amount of play at the wheel when it is being adjusted in the vise.) Moving to the top set of cone/locknut, I use the cone wrench on the cone and turn it to where there is just the tiniest amount of play which I feel by holding the rim and rocking it up and down. I lock the cone and locknut together in the jambnut fashion to keep the adjustment.

In effect, moving the cone up puts less pressure on the bearings. Moving it up is done by "unscrewing" it. One of the finer points of jambnutting is that it can be done by moving the top nut down or the bottom one (the cone in this case), up or kind of moving them into each other. The important thing is to achieve the correct adjustment and have the locknut and cone locked together tightly enough that they won't move when the wheel is turning.

A cone that is not locked securely to the lock nut may turn itself inward and put pressure on the bearings which is el desastor!

The reason we leave the tiny amount of play in the bearing is that the axle is squoze (ha!) veddy veddy tightly by the quick release lever. That pressure actually bends the axle enough to change the setting that felt just right with no play into one that can put a fair amount of pressure on the bearings. If you would like a demonstration of this effect, come into the shop and if I'm not busy, I'll show you how that works. By the way, there is no need for play on nutted hubs.

With the exception of leaving play to compensate for the pressure of the quick release skewer, that is how the bottom bracket and headset are adjusted on old school bikes.

Also, the same principle applies to the center bolts of side pull brakes.

I will continue this if it seems that people want to learn it and have the patience to read it.

Cheers, Chuck

2 comments:

  1. Chuck,
    I like the VW valve adjustment analogy.

    I hope this doesn't sound like a stupid question: Is the hub adjusted from one side only or both?

    If you had to put a measurement (in inches or mm) for the play needed with QR hubs, what would you say it should be?

    On threadless headsets, I don't understand how the star nut works or why it's needed. Doesn't the stem do all the work holding the steerer? Again, please forgive me if this seems like another stupid question.

    ReplyDelete
  2. Fatmex, thanks for the good questions.

    On the hub, the axle is one piece with two ends. Both ends are adjustable but in practice you jamb nut one end (lock it) and then do the adjustment from the other end. It doesn't matter which side you lock and which side you choose to adjust. If one side would be easier for some reason such as a freewheel or access, it is only important that both sides are locked. The distance between the cones is what you are adjusting.

    Play in QR hubs is just a tad. If you have a vise and the wheel is held by the axle or locknut, you should be able to just percieve play at the rim. I check that thusly: left hand index finger and thumb holding the top of the axle (Remember it is suspended horizontally in the vise); right hand out at the rim, move the rim up and down, which is equivelent to sideways if it was on the bike. You should feel just a tiny bit of play. Final test is that the play should be taken out when the QR is tightened on the bike. Ideally, the wheel will swing back and forth easily when the valve, which is the heavy point, comes to and passes the bottom.

    The star nut is there only to facilitate taking out the play from the headset. You are correct that the stem does the work of holding that adjustment. The same thing could be accomplished without the star nut. There would need to be some other method,for instance a clamp, temporarily used to snug the play out of the bearings, or a threaded rod, some washers and a couple of nuts could be used inside the steerer and then removed after you tighten the stem to hold the correct adjustment. Note: even though those methods would work, they are given in the hopes that they clarify the job done by the star nut.

    Chuck

    ReplyDelete